Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    phrases related to physical actions or events, particularly those involving mistreatment or shocking behaviors

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    orbit
    -0.90
    Strange
    -0.89
    é¾
    -0.89
    mare
    -0.88
    Introduced
    -0.88
    operated
    -0.88
    fleet
    -0.88
    illin
    -0.87
    heart
    -0.86
    glass
    -0.86
    POSITIVE LOGITS
     example
    1.58
     instance
    1.53
    geries
    1.49
    bidden
    1.36
     purposes
    1.35
    gery
    1.29
     starters
    1.28
     reasons
    1.26
     awhile
    1.26
    gotten
    1.25
    Act Density 2.404%

    No Known Activations